Download 34. Sampling, shipping, organisation and testing of inactivated samples to take advantage of molecular analysis Methods

The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008
Appendix 34
SAMPLING, SHIPPING, ORGANISATION AND TESTING OF INACTIVATED SAMPLES TO
TAKE ADVANTAGE OF MOLECULAR ANALYSIS METHODS
*1+
S. Alexandersen
1
1
1
2
, J. Klein , T. Frederiksen and M. Hussain .
Department of Virology, National Veterinary Institute, Technical University of Denmark, Lindholm,
DK-4771 Kalvehave, Denmark
+
2
Affiliation from 1 October 2008, National Centre for Foreign Animal Disease, Canadian Food
Inspection Agency, 1015 Arlington Street, Winnipeg, R3E 3M4 Manitoba, Canada
Food and Agriculture Organization of the United Nations - Pakistan, NARC, Islamabad, Park Road,
PK-45500, Pakistan
ABSTRACT
Introduction
FMD is endemic in many countries and causes significant economic losses in livestock farming
despite efforts directed towards control by vaccination. Efficient vaccination for protection against
FMD in such settings is often hampered by a limited access to or excessive costs of high quality
vaccines and/or the lack of a well planned vaccination strategy based on thorough epidemiological
information. Limited access to or excessive costs of high quality vaccines is a matter of economic
resources and may be solved by sufficient support from International Organisations. However, in
such settings the methodological tools for generating thorough epidemiological information and
sufficient and timely characterisation of circulating and epidemiological important strains of FMDV
required for setting up a well planned vaccination strategy need a leap forward in regard to the
ways epidemiological studies and strain characterisation are organised and performed. Previous
experimental studies performed by us and others have looked at new sampling strategies for
relatively easy estimation of e.g. prevalence of infected animals by using real-time RT-PCR analysis
of swab samples collected directly into a suitable lysis buffer. The next step was to look at this
approach under field conditions by using sampling of mouth swabs directly into a suitable lysis
buffer for easy shipping and downstream analyses as described here.
Materials and methods
For this study we selected the Landhi Dairy/Cattle Colony (LCC) outside Karachi, the largest dairy
colony in Pakistan and the biggest buffalo colony in the world. Due to its size and fragmented
vaccination coverage, together with the Pakistani husbandry tradition of frequent transportation of
animals to and from the large dairy colonies, FMDV infection is prevalent in LCC and therefore, LCC
provided a suitable study population. Moreover, the information generated would also provide
epidemiological information that could potentially improve the control of FMD in the colony and
thus be of benefit to the many farmers enduring significant economic losses. From April 2006 to
April 2007 we collected mouth swab samples of randomly selected, non-clinically affected herds as
one group (group A), from non-clinically affected animals in herds with clinical evidence of a prior
FMD outbreak (group B, i.e. herds with a few animals with evidence of old and healing lesions
consistent with acute FMD for more than 2-3 weeks prior to sampling) and from non-clinically
affected animals in a few herds in which acute FMD was obvious in at least one animal (group C).
Swabs were immediately placed in Qiagen RLT buffer, which immediately stabilised the viral RNA
and rendered the sample non-infectious. In addition, epithelial samples were collected from clinical
FMD cases from group C. The results from analysis of these epithelial samples are described in a
separate paper together with the detailed sequence and phylogeny studies of selected positive
mouth swab samples from all 3 groups. The mouth swab sampling scheme included four major
samplings together with smaller monthly samplings. Each sampling was accompanied by a
questionnaire to gather general health status and vaccination information and the exact location of
the herd was determined by GPS. The laboratory-based analysis employing real-time RT-PCR is
described here
203
The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008
While VP1, capsid and full genome sequences derived from swabs and epithelial samples, together
with the questionnaire-based information, are described in an accompanying paper.
Results
Sampling of mouth swabs directly into lysis buffer followed by shipment of the samples either
directly to Denmark for real-time RT-PCR analysis at Lindholm, or alternatively via the National
Reference Laboratory in Islamabad for RNA extraction followed by cDNA preparation and shipping
of prepared cDNAs to Denmark, proved to be a reliable method for detecting FMDV RNA. Sampling
in this way preserved the FMDV RNA and such samples contain no infectious FMDV and can
therefore relatively easy be shipped as diagnostic samples and clearly provided valuable
information on prevalence of FMDV infection in LCC over the one-year study period. Moreover,
sequencing of such samples were possible provided that they contained a reasonable amount of
FMDV RNA (sequencing results to be described in the accompanying paper) and in addition,
preliminary in vitro results (Graham Belsham, personal communication) indicate that it may be
possible to re-generate infectious FMDV from such purified RNA samples by transfection into
suitable cell cultures.
Discussion
The approach described here has in our opinion clearly shown merit. Stabilised and inactivated
swab samples can easily be collected and shipped internationally and may be used to determine
the temporal-spatial distribution and prevalence of FMDV infection and further analysis of selected
positive samples by sequencing can be used to characterise circulating strains of FMDV. Thus, the
described method is a viable and valuable alternative to shipment of epithelial samples, containing
infectious virus, for strain characterisation.
1. INTRODUCTION
Pakistan has a number of large dairy buffalo and cattle colonies, FMD is endemic there and it has
borders with India, Afghanistan, Iran and China where FMD is also a continued problem. Only
fragmented information exists on the prevalence and extent of circulation of FMDV, including
characterisation of current strains, in Pakistan. Improvement of this situation requires intensified
surveillance and facilities and expertise in molecular epidemiology and may initially best be
achieved by international collaboration. The project described here is such a collaboration between
experts and staff in Islamabad and Karachi in Pakistan and at Lindholm in Denmark in close
coordination with the FAO and the WRL for FMD in Pirbright, UK.
The traditional diet in Pakistan includes buffalo milk or buffalo milk-based products and large dairy
colonies are located around the major cities like e.g. Karachi to provide a continued supply of fresh
non-pasteurised milk. Buffalo milk is preferred over cattle milk due to the high fat content and
many of the large dairy colonies have 90-95% buffalo and only 5-10% cattle. The Landhi
Dairy/Cattle Colony (LCC) is located in the eastern suburbs of Karachi, was established in 1959 and
is now the largest dairy colony in Pakistan and the biggest buffalo colony worldwide. There is
estimated to be a total of around 250-280 000 buffalo and 15-25 000 cattle in the colony. The
number of individual farms is not known with certainty, but it has been estimated that there is
between 1000-2000 farms each with around 100-200 animals on average. All of this is located on
only 750 acres of land (including 250 acres for primitive roads). The farmers mainly purchase
animals from the three major breeding regions in Punjab and in upper and lower Sindh and animals
are primarily kept for one lactation period. After the lactation period, a few of the buffalos/cattle
are used for re-breeding by the owner, some are slaughtered, and an estimated 50-75% of them
go back to the regions mentioned above to be used for breeding once more. Such animals will,
when they are ready to enter a new lactation period, again be transported across the country and
sold to individual farmers.
Many goats, sheep, cattle and buffalo from all over Pakistan are once a year transported to large
markets just outside the big cities in connection with the celebration of the major Muslim festival
Eid-ul-Azha. Animals that are not sold for sacrifice may, after having been mixed with many other
animals from different places, then go back to where they came from or perhaps be sold and go
somewhere else.
The majority of the commercial farmers in LCC vaccinate their animals, either with a good quality
trivalent vaccine (O, A, Asia1), but only given once due to the high cost of the vaccine, or with a
monovalent type O vaccine produced in Pakistan which is of unknown quality and efficiency and
apparently only contains antigen from a single, very old type O strain. This lack of an efficient and
coordinated vaccination policy, together with the large number of animals kept in a relatively small
area and the frequent transportation of animals in and out of the colony as well as the general lack
of biosecurity procedures and awareness among LCC farmers, creates conditions that favour
204
The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008
continued FMDV circulation. Therefore, the LCC provides a suitable study population for
establishing the methods needed for direct inactivation, stabilisation and shipment of field samples
containing FMDV RNA and such studies may in turn provide useful epidemiological information to
be used to improve FMD control in LCC.
In order to potentially set up a future vaccination strategy that is efficient in protecting the LCC
against currently circulating strains of FMDV, it is necessary to get data on the seasonal occurrence
and prevalence of FMDV in the colony and to characterise the circulating strains. This may be
achieved by a structured sampling strategy based on mouth swab sampling in connection with the
relatively easy shipment of inactivated and stabilised samples to a laboratory with the capacity to
perform large-scale RT-PCR testing to detect samples positive for FMDV which can then
subsequently be further analysed by sequencing as described below and in an accompanying
paper.
2. MATERIALS AND METHODS
2.1 Study Design and Sample Collection, Organisation and Shipping.
The study was designed as a longitudinal and cross-sectional survey to provide information on
clinical and in particular subclinical FMDV infection prevalence in LCC. We collected samples during
4 trips to Pakistan from April 2006 to April 2007 and local Veterinary Officers collected in addition a
smaller number of samples each month during the same period. The monthly smaller samplings
included 5 farms selected at random and on each farm 6 animals (5 buffalo and 1 cow if possible)
were selected at random, clinically examined and a mouth swab taken by careful swabbing of the
tongue with a sterile cotton swab (Libby Sterilin). During our 4 visits, the number of randomly
selected farms was increased to 18 and the number of randomly selected animals on each herd
increased to 9 buffalo as well as 1 cow if possible. When visiting the farms, the location was logged
by handheld GPS equipment and the farmers thoroughly questioned for knowledge that may be of
relevance for the analysis, e.g. clinical signs, vaccination status, vaccine used, age of the animal,
outbreak history and any other relevant information. The randomly selected animals were checked
clinically while taking samples. Farms without any clinical signs of current or relatively recent acute
FMD were assigned to study group A (randomly selected, non-clinically affected animals from nonaffected herds). Animals from farms with a history and signs of recent clinical FMD were assigned
to study group B (samples were taken from non-clinically affected animals in herds where the
presence of a few other animals with evidence of old and healing lesions indicated that acute FMD
had been present in the herd for more than 2-3 weeks before sampling) and farms with animals
with current signs of acute FMD found during the initial examination were assigned to study group
C (samples were taken from non-clinically affected animals in herds in which acute FMD was
obvious in at least one other animal). The animals in the herds of group B and C with clinical signs
of earlier or current FMD were examined in more detail, in particular concerning animals in group C
in order to locate suitable vesicular lesion epithelia (lesion epithelia samples were in addition also
collected during a pre-study visit in January/February of 2006; the results of the testing of
collected epithelial samples are described in an accompanying paper). Mouth swabs were
immediately placed, after breaking and removing the plastic handle of the swab, into 1.0 ml of RLT
lysis buffer (Qiagen) in 2 ml screw-cap polypropylene tubes (Sarstedt). Such swabs in lysis buffer
o
were often kept at ambient temperature for many hours before being stored at -20 C and then
shipped either directly to Denmark, or alternatively via the National Reference Laboratory in
Islamabad for RNA extraction followed by cDNA preparation and shipping of prepared cDNAs to
Denmark. Swabs in lysis buffer and prepared cDNAs as described here contain no infectious FMDV
(only stabilised RNA or cDNA) and can relatively easily be shipped as diagnostic samples both
locally and internationally.
2.2 RNA Extraction, cDNA Preparation and Quantitative RT-PCR
Total RNA was extracted from thawed sample aliquots using the Qiagen RNA Blood Kit (Qiagen)
according to the manufacturer's instructions. The extracted RNA samples were then converted into
cDNA and analysed by quantitative “real-time” RT-PCR to determine the amount of FMDV RNA
present in the mouth swab samples as described in detail elsewhere, with the assumption that the
swabs were initially diluted around 10-fold when placed in 1 ml of RLT lysis buffer
2-4,10-12,15
.
Sequencing. Selected samples with a relatively high content of FMDV cDNA based on the RT-PCR
result were further analysed by sequencing essentially as described by us previously
results of the sequencing are described in an accompanying paper.
6, 7
. The
205
The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008
3. RESULTS
3.1 FMDV RNA Detected in Swab samples.
During the 1 year study we collected 960 mouth swabs from 124 farms in group A (herds with no
clinical signs of acute or recent FMD), 45 samples from 4 herds in group B (animals without clinical
signs, but from herds where other animals had signs of healing/healed FMD) and 29 samples from
2 herds in group C (animals without clinical signs, but from herds where other animals had acute
clinical signs of FMD) giving a total of 1034 swab samples analysed. The number of FMDV RNA
positive swabs was 106 in group A (no clinical signs), 22 in group B (herds with relatively recent
FMD) and 25 in group C (herds with acute signs of FMD), corresponding to 11%, 49% and 86% of
the samples being positive in groups A, B and C, respectively. Of the 106 positive samples
collected from subclinically infected animals/herds in group A, 58 contained sufficient FMDV RNA to
facilitate subsequent sequencing. This selected group of 58 positive swabs from subclinically
infected animals had a mean Ct value in the real-time RT-PCR of 36.0 (range 26 to 39) equivalent
4.6
3.7-7.5
) FMDV RNA target copies per ml of saliva/swab while the other 48
to around 10 (range 10
swabs from this group, not selected for sequencing, had a mean Ct of 42.5 (range 40-48)
2.7
1.0-3.4
equivalent to around 10
(range 10
) copies per ml of saliva/swab. For comparison, the
positive swabs from the farms with recent signs of FMD had a mean Ct of 37.5 (range 25-46)
equivalent to around 10
4.1
1.6-7.8
(range 10
) copies per ml of saliva/swab while the positive swabs
5.8
collected in farms with acute FMD had a mean Ct of 32.0 (range 24-39) equivalent to around 10
(range 10
2
3.7-8.1
10 (range 10
) copies per ml of saliva/swab. These copy numbers usually correspond to roughly
1-5
1
), 10 (range 10
0-1.5
1.5
), 10
(range 10
0-5
3
) and 10 (range 10
1-5.5
) tissue culture
1-3,10
infectious doses (TCID50) per ml when assayed in bovine thyroid cells
. For further details of
seasonal prevalence of subclinical infection and sequence analysis etc please see the accompanying
paper and our previously published results
6,7
.
4. DISCUSSION
The data presented point out that carefully planned field epidemiological studies combined with
collection of swab samples directly into a suitable lysis buffer for subsequent easy shipping, locally
or internationally, to a laboratory for molecular testing by RT-PCR followed by sequencing of
selected samples may provide a powerful tool for generating epidemiological information and
valuable characterisation of circulating strains of FMDV. More than 10% of samples collected from
herds with no signs of FMD turned out to be positive for FMDV RNA, while herds having animals
with signs of an earlier outbreak of FMD (healed lesions) had approximately 50% of the samples
from apparently unaffected neighbouring animals being positive and more than 80% of unaffected
animals were positive in herds where other animals had signs of acute FMD. Taken together, these
data indicate that the method described provides an accurate account of the numbers of animals
being exposed to FMDV within a relatively short time period prior to sampling and in fact, may
indicate that the positive animals are indeed infected. Data from our experimental studies indicate
that mouth swabs are usually detected as positive within a time window of up to 2 weeks after
2, 3
exposure and are negative in carrier animals
. Consequently, we consider it likely that the
positive animals detected here in a field setting were exposed or infected within a 14 days window.
Although FMDV RNA positive animals without clinical signs of disease may not for certain be
infected, we consider the fact that more than 50% of the positive samples from farms with no
3.7-7.5
clinical signs of infection contained enough RNA sufficient for sequence analysis (a level of 10
1-5
FMDV RNA target copies corresponding to roughly 10 TCID50 per ml of saliva/swab) an indication
that these animals may be subclinically infected. This notion is substantiated by the additional facts
that the RT-PCR targets the 5’-untranslated region (5’-UTR) and that the sequencing PCRs involved
amplification of relatively long targets, which together have been shown to be good indicators of
picornavirus infectivity
5,8,9,13,14
, and that our previous studies have indicated a good correlation
1-3,10
between copy numbers and infectivity
; i.e. further evidence for these animals indeed being
subclinically infected and that FMDV likely replicated
in these animals to levels facilitating detailed molecular analysis. This view is further supported by
the finding that non-affected animals in farms with other animals having evidence of previous, but
healed, FMD lesions had a higher percentage of swabs positive for FMDV RNA, indicating an earlier,
but apparently relatively widespread, circulation of virus in this group and potentially still
206
The Global control of FMD - Tools, ideas and ideals – Erice, Italy 14-17 October 2008
subclinical infection in some animals. Also, the finding of a very high percentage of positive swabs
and a higher RNA content in swabs from non-affected animals from herds with ongoing acute FMD,
further substantiates that such animals may truly be subclinically infected or alternatively, be in an
early stage of infection before lesions develop. Consequently, although some of the animals
detected as FMDV RNA positive by our methods may only have been exposed to FMDV and not
necessarily actively infected, we consider it likely that a significant proportion was indeed
subclinically infected.
The use of on-site/in-the-field inactivated and stabilised swab samples as described here for easy
shipping and subsequent analysis in advanced international laboratories, may also be combined
with collection of epithelial samples from clinically affected animals as such epithelial samples also
can be placed directly into a suitable solution, e.g. RNAlater (Ambion), which inactivates viral
infectivity and stabilises the RNA provided that only small samples are collected and that the
tissues are fully immersed in the RNAlater and kept at room temperature for at least 24 hours
before shipping (Alexandersen, S. et al. unpublished). Finally, preliminary in vitro results (Graham
Belsham, personal communication) indicate that, provided collected samples contain sufficient
FMDV RNA, it may likely be possible to re-generate infectious FMDV from such RNA purified from
inactivated and stabilised swabs or epithelia by using transfection into suitable cell cultures should
the need arise to study the virus further for e.g. virulence or vaccine studies etc.
5. CONCLUSIONS
Combined field and molecular epidemiological studies using randomly selected mouth swabs
collected directly into a suitable lysis buffer provide a powerful tool for generating detailed
epidemiological information and molecular characterisation of circulating strains.
Direct lysis of collected swab samples into a suitable lysis buffer circumvents the need for
international shipment of samples as dangerous goods and provides inactivated and stable
samples that without any safety problems can be shipped as diagnostic samples.
Analysis of high numbers of mouth swabs by a highly sensitive, quantitative real-time RT-PCR
indicates that this method may provide an efficient way of establishing the extent of virus
circulation during periods of low clinical activity and for providing estimates of prevalence of
infection at a given point in time. Furthermore, subsequent sequencing of selected samples can
be used for detailed analysis of the relationship of circulating strains with strains circulating in
other countries or with known vaccine strains in order to provide appropriate vaccine coverage.
Extracted RNA may possibly be used for making infectious FMDV by using transfection into
suitable cell cultures; however, this aspect needs further study.
6. RECOMMENDATIONS
Collaborative studies using inactivated and stabilised samples and involving the relevant
authorities and laboratories of countries with endemic FMD, together with International
Organisations such as FAO, the WRL for FMD and a National Reference Laboratory from a
resource-rich country may provide an efficient avenue for strengthening FMD control programs.
Further studies should be supported to further establish the suitability of the proposed methods,
to provide more epidemiological data and knowledge of circulating strains of FMDV in various
settings and to establish the potential for using such samples to re-generate infectious FMDV by
optimising the methods used for lysis and stabilisation, RNA extraction and in particular the
methods for maximising the efficiency of transfecting susceptible cells.
7. ACKNOWLEDGMENTS
We thank the many friendly and competent Veterinary Officers that have helped us in Pakistan.
Drs. Giancarlo Ferrari and Keith Sumption from FAO are thanked for their continued interest in the
project. Finally, we thank The National Veterinary Institute, Technical University of Denmark, the
FAO Regional Project (GTFS/INT/907/ITA) and the EU Network of Excellence for Epizootic Disease
Diagnosis and Control (EPIZONE, Call Identifier: FP6-2004-Food-3-A) for support.
8. REFERENCES
[1] Alexandersen, S. and N. Mowat. 2005. Foot-and-mouth disease: host range and
pathogenesis. Curr. Top. Microbiol. Immunol. 288:9-42.
207
The Global control of FMD- Tools, ideas and ideals – Erice, Italy 14-17 October 2008
[2] Alexandersen, S., M. Quan, C. Murphy, J. Knight, and Z. Zhang. 2003. Studies of
quantitative parameters of virus excretion and transmission in pigs and cattle experimentally
infected with foot-and-mouth disease virus. J. Comp Pathol. 129:268-282.
[3] Alexandersen, S., Z. Zhang, A. I. Donaldson, and A. J. Garland. 2003. The Pathogenesis
and Diagnosis of Foot-and-Mouth Disease. J. Comp Pathol. 129:1-36.
[4] Alexandersen, S., Z. Zhang, S. M. Reid, G. H. Hutchings, and A. I. Donaldson. 2002.
Quantities of infectious virus and viral RNA recovered from sheep and cattle experimentally
infected with foot-and-mouth disease virus O UK 2001. J. Gen. Virol. 83:1915-1923.
[5] Bhattacharya, S. S., M. Kulka, K. A. Lampel, T. A. Cebula, and B. B. Goswami. 2004. Use
of reverse transcription and PCR to discriminate between infectious and non-infectious hepatitis A
virus. J Virol. Methods 116:181-187.
[6] Klein, J., M. Hussain, M. Ahmad, M. Afzal, and S. Alexandersen. 2008. Epidemiology of
foot-and-mouth disease in Landhi Dairy Colony, Pakistan, the world largest Buffalo colony. Virol. J
5:53-68.
[7] Klein, J., M. Hussain, M. Ahmad, P. Normann, M. Afzal, and S. Alexandersen. 2007.
Genetic characterisation of the recent foot-and-mouth disease virus subtype A/IRN/2005. Virol. J.
4:122-133.
[8] Li, J. W., Z. T. Xin, X. W. Wang, J. L. Zheng, and F. H. Chao. 2002. Mechanisms of
inactivation of hepatitis a virus by chlorine. Appl. Environ. Microbiol. 68:4951-4955.
[9] Li, J. W., Z. T. Xin, X. W. Wang, J. L. Zheng, and F. H. Chao. 2004. Mechanisms of
inactivation of hepatitis A virus in water by chlorine dioxide. Water Res. 38:1514-1519.
[10] Quan, M., C. Murphy, Z. Zhang, and S. Alexandersen. 2004. Determinants of early footand-mouth disease virus dynamics in pigs. J. Comp Pathol. 131:294-307.
[11] Reid, S., N. Ferris, G. Hutchings, Z. Zhang, G. Belsham, and S. Alexandersen. 2002.
Detection of all seven serotypes of foot-and-mouth disease virus by real-time, fluorogenic reverse
transcription polymerase chain reaction assay. J. Virol. Methods 105:67-80.
[12] Reid, S. M., S. S. Grierson, N. P. Ferris, G. H. Hutchings, and S. Alexandersen. 2003.
Evaluation of automated RT-PCR to accelerate the laboratory diagnosis of foot-and-mouth disease
virus. J. Virol. Methods 107:129-139.
[13] Simonet, J. and C. Gantzer. 2006. Degradation of the Poliovirus 1 genome by chlorine
dioxide. J Appl. Microbiol. 100:862-870.
[14] Simonet, J. and C. Gantzer. 2006. Inactivation of poliovirus 1 and F-specific RNA phages
and degradation of their genomes by UV irradiation at 254 nanometers. Appl. Environ. Microbiol.
72:7671-7677.
[15] Wernery, U., P. Nagy, C. M. Amaral-Doel, Z. Zhang, and S. Alexandersen. 2006. Lack of
susceptibility of the dromedary camel (Camelus dromedarius) to foot-and-mouth disease virus
serotype O. Vet. Rec. 158:201-203.
208